Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    names with a specific ending pattern

    oai_token-act-pair · gpt-3.5-turbo

    the word "oud" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    INAL
    -0.77
    åī
    -0.75
    ASE
    -0.74
    UG
    -0.72
    RECT
    -0.65
    ANC
    -0.65
    =-=-=-=-
    -0.64
    Forge
    -0.64
    BOOK
    -0.64
    UNCH
    -0.63
    POSITIVE LOGITS
    oud
    1.26
    reau
    1.05
    emouth
    1.02
    ecided
    1.00
    oun
    0.99
    escription
    0.99
    spe
    0.94
    iosyncr
    0.94
    emo
    0.88
    mond
    0.87
    Act Density 0.005%

    No Known Activations